questions for 8.a.i Investment Risk (Grinold)

[ajsa]

Hi David,

I got some questions after viewing the screencast:

1. How to handle "lack alpha forecast for stocks in benchmark"?

2 "The size of the no-trade region depends on the transactions costs, the risk aversion, and the expected return and risk of stocks and bonds." But I think the size of the region just depends on the tran cost (purchase cost + sell cost).. could you clarify?

3. why in MCVA formula, there is a 2 in the 2nd term? It looks inconsistent with the fomular of ValueAdded.

4. Should ValueAdded consider/exclude transaction cost? Is the puropse of port construction to maximize ValueAdded or alpha?

5. Schweser says "high risk aversion and low transaction cost leads to low TE. And without transaction cost, there will be no TE or dispersion". Does it make sense? Could you explain?

6. One methods to reduce dispersion to zero is to "invest the composite in the new optimum.
–But requires paying excess transactions costs.
–By treating existing portfolio and new portfolio separately, accepts some level of dispersion for higher returns."
Do you mean the "composite" is the new portfolio or the whole composite? If it is new portfolio, why it requires paying excess transactions costs? if it is the whole composite, why does it "treat existing portfolio and new portfolio separately"?

Thanks!

 

[hsuwang]

Hello David,

I felt a bit confused specifically with the chapter on Portfolio Construction as well. I read the Schweser notes and watched your video and I think both didn't really give too much explanation on the formulas (No-trade region, MCVA, risk aversion). My take on this is maybe just memorize the formula instead of really digging into where they come from and what they mean? The AIMs did not specify there's a calculation for, say, MCVA or the no trade region, so it is probably safe to assume the formulas won't be tested?

also you mentioned in the video (8.a.ii @13:17) alpha = residual risk, but isn't alpha = residual return and residual risk = TE of alphas?


Thanks!

 

[David Harper CFA FRM]

@Jack: I agree with you. A challenge with both the assigned Grinold Ch 7 & 17 is that they basically (mostly) assume you've read the book up until then; so the formulas appear out of thin air. That's why I did some "Grinold setup," to share some of the prior foundation...
I totally agree with you: as a practical matter, (and i think i mentioned this is the video), the testability of Ch 17 is extremely low...I don't think I've seen a single "real" question on it... so when you say, "My take on this is maybe just memorize the formula instead of really digging into where they come from and what they mean?" I 100% agree ... from an exam-taking strategy, I would *not* spend a whole lot of time here ... the XLS spreadsheet I built that that decomposes active systemic return summarizes most of what you'd want to know, but I doubt even that may will be tested (in 2009) at all...

Re 8.a.ii: oops, my typo! That is an error on the slide, my apologies, i will add to the errata...yes, of course you are correct that "alpha = residual return and residual risk = TE of alphas?"

@asja: I will return to your questions when I have time to do the research...but as they aren't "mission-critical" to the exam, please forgive a delay...

Thanks, David